OBJECTIVES: 1. To determine the effect of anticonvulsant prophylaxis on the incidence of post-traumatic epilepsy. 2. To compare the effect of anticonvulsant prophylaxis on the EEG recordings of untreated and treated patients. 3. To determine the accuracy of the presence or absence of PLEDS for prediction of the development of post-traumatic epilepsy. 4. To attempt to beter define brain injury factors that contribute to the development of post-traumatic epilepsy. METHODS: 1. Randomization of 200 head injured patients susceptible to developing post-traumatic epilepsy into control and treated groups. 2. Treated groups receive phenytoin according to attached protocol. 3. Control and treated groups are statistically compared with regard to the incidence of seizures and differences in EEG patterns. 4. Patients developing and not developing post-traumatic epilepsy are statistically compared with regard to: a) findings on computed tomography brain scans; b) serial levels of lactic dehydrogenase isoenzymes and c) performance on psychometric tests. Experimental evidence suggests post-traumatic epilepsy can be prevented by anticonvulsant prophylaxis. A well-controlled study evaluating the pharmacologic prophylaxis of post-traumatic epilepsy has not been published. This lack of statistical data is a major reason standardized indications and methods for early seizure prevention have not been established. This randomized study of 200 patients is proposed to statistically analyze the effect of anticonvulsant prophylaxis on the incidence of post-traumatic epilepsy and to better define the brain injury factors contributing to the development of post-traumatic epilepsy.